Spatial and temporal scales of variability for indoor air constituents

The study addresses whether indoor air constituents are well mixed or display significant spatial and temporal variability, particularly during common activities like bleach cleaning. Prior assumptions of homogeneous mixing underlie many single-point measurement strategies and box-model approaches, but may not hold for reactive and short-lived species such as radicals and bleach-derived products. With people spending about 90% of their time indoors and with higher indoor concentrations driven by activities like cleaning and cooking, understanding spatial distributions is critical for exposure assessment. Bleach use emits HOCl and Cl2 and, via multiphase reactions with nitrite and ammonia in applied solutions, forms chlorinated and nitrogenated species (e.g., ClNO2, NCl3) that can be hazardous. The purpose is to quantify spatial and temporal variability of these species by combining detailed chemical models and CFD, and to evaluate implications for indoor exposure and indoor–outdoor transport.

Previous work has often assumed indoor spaces are well mixed, applying box models with deposition velocity concepts that treat a uniform core region and surface boundary layers. Measurements resolving spatial or vertical gradients are relatively rare. CFD has been used primarily for non-reactive species to resolve airflow and distributions. The HOMEChem campaign revealed multiphase bleach chemistry producing HOCl, Cl2, ClNO2, and chloramines through interactions with nitrite and ammonia on surfaces and in applied solutions. Prior observations during bleach cleaning showed enhanced indoor chlorine chemistry and formation of toxic compounds. These findings suggest that for reactive and short-lived species, well-mixed assumptions may be invalid, motivating the integrated modeling approach here.

The study integrates three modeling frameworks with extensive measurements from the HOMEChem campaign: (1) a multiphase kinetic model, (2) the INDCM detailed photochemical box model, and (3) a CFD model. The HOMEChem bleach experiment occurred on June 8, 2018, in a 111 m², 3-bedroom, 2-bath test house (Austin, Texas). As part of a layered experiment (cooking prior to bleaching), a bleach solution was applied to kitchen and living room floors (combined 40 m²) at 17:35. Gas-phase concentrations were measured before and during the experiment using two TOF-CIMS instruments (iodide CI for HOCl, Cl2, ClNO2, NCl; acetate CI for HONO), a cavity ring-down spectrometer (NH3), and a LIF-FAGE instrument (OH, HO2, HONO) next to the living room window. Spectrally resolved solar irradiance was co-located with LIF-FAGE. The HVAC-controlled air exchange rate during the experiment was ~0.7 h⁻1. The multiphase kinetic model treated outdoor-indoor exchange, gas-phase reactions, photolysis, wall loss, heterogeneous uptake to surfaces and particles, aqueous bleach reactions, and transport of semi-volatile species through a boundary layer above the bleach surface. It produced time-varying concentrations of HOCl, ClNO2, chloramines (e.g., NCl3), and NH3 directly above the bleach surface. INDCM (a near-explicit mechanism) quantified radical production and OH reactivity, including photolysis, exchange, internal emissions, and deposition. The CFD geometry mimicked the test house airflow and emission conditions and resolved turbulent airflow, solar radiation through windows, surface uptake, and a reduced chemical mechanism of 11 reactions. Because full chemistry is computationally prohibitive in CFD, key inputs from the detailed models constrained the CFD: INDCM-derived OH production rates and reactivity; multiphase-model-derived time-dependent surface emissions of HOCl, ClNO2, NCl3, and NH3; and critical photolysis rates, rate and uptake coefficients. Measurements were primarily at point P2 (kitchen) for most species and at P7 (sunlit window zone) for OH, with nine CFD sampling points at 1.5 m height.

• The integrated modeling (multiphase kinetic, INDCM, CFD) reproduced the temporal evolution and spatial variability of OH, HOCl, NCl3, and NH3 observed during bleach cleaning. The INDCM captured OH concentrations; the kinetic model matched HOCl, NCl3, and NH3 time series; the CFD reproduced spatial heterogeneity at sampling points. - Mechanistic pathway: OH enhancement during cleaning is mainly driven by a cascade initiated by Cl2 photolysis. Cl radicals react with VOCs producing peroxy/alkoxy radicals that propagate to HO2 and then OH via NO chemistry. Gas-phase simulations attribute >90% of OH production to this chain; additional OH arises from HOCl and HONO photolysis. - OH reactivity during cleaning was high, ~65 s⁻1, indicating rapid OH loss to NOx and VOCs. - Spatial patterns: High OH was confined to sunlit zones near windows; low in dark zones due to rapid loss. HOCl, ClNO2, and NCl3 exhibited vertical gradients, with higher concentrations in the living room where bleach was applied. Despite an AHU mixing rate of 8 h⁻1 and open doors, products remained concentrated in the living room and were further localized near a corner (P1), with 30–50% higher concentrations than elsewhere due to non-uniform airflow. Cl2 showed similar spatial distributions. NH3 was relatively homogeneous, with a few ppb depletion in the cleaning area due to uptake and aqueous reactions. - Temporal and spatial scale framework (for typical AER 0.5 h⁻1 and indoor conditions; airflow ~0.03 m s⁻1): (1) Microscale (<~0.1 m; lifetimes up to ~10 s): radicals (Cl, NO3, ROx) show sharp gradients governed primarily by reaction rates. (2) Room scale (~0.1–10 m; ~10 s–10 min): moderately long-lived species (NH3, NO, Cl2, O3) have spatial gradients controlled by chemistry and airflow; for NH3 and SVOCs, reversible surface partitioning reduces apparent gradients via surface emissions. (3) Building scale (>~10 m): long-lived species (VOCs, NO2, CO2) are mostly well mixed; temporal scales controlled by ventilation. - Particles: 100 µm settle to the floor within seconds and ~1 m (gravitational settling). Ultrafine particles (1–10 nm) are short-lived due to diffusion losses and coagulation. Particles of 1–10 µm persist minutes to ~1 h, comparable to ventilation timescales, enabling transport to breathing zones and potential roles in airborne pathogen transmission (e.g., SARS-CoV-2) and as carriers of thirdhand smoke species. - The analysis implies SVOCs generated indoors can be transported outdoors, influencing urban air quality depending on reactivity and ventilation.

The findings challenge the prevailing well-mixed assumption for indoor environments by demonstrating significant horizontal and vertical gradients for short- and moderately long-lived species produced during bleach cleaning. By integrating detailed chemical models with CFD, the study shows that chemical reactivity, photolysis driven by spatially heterogeneous irradiance, surface interactions, and ventilation jointly determine indoor concentration fields and decay rates. Confinement of OH to sunlit areas and localization of bleach products near emission zones highlight the importance of considering micro-environmental exposure hotspots rather than relying on single-point measurements. The framework of spatial and temporal scales clarifies when box-model approaches may be inadequate, particularly for reactive species where deposition velocity concepts may need revision. The results have implications for human exposure assessments, for indoor-to-outdoor transport (e.g., VOCs and SVOCs influencing ambient ozone and secondary organic aerosol), and for interpreting measurements from campaigns like HOMEChem under varying ventilation conditions.

Heterogeneous spatial distributions of indoor pollutants can exist for short- and moderately long-lived compounds during common activities like bleach cleaning, counter to traditional homogeneous mixing assumptions. Spatial and temporal variability is governed by gas- and multiphase chemistry, photolysis, deposition, airflow, and outdoor–indoor exchange. The integrated modeling approach reconciled complex chemical processes and airflow to reproduce observations, revealing OH production largely initiated by Cl2 photolysis and substantial room-scale gradients in bleach-derived species. The work underscores the need to better characterize surface interactions—affected by temperature, humidity, light, surface pH, and organic films—to improve predictions of indoor air quality, exposure, and indoor–outdoor chemical transport. Future research should resolve boundary-layer and surface processes more explicitly, expand chemical mechanisms within CFD frameworks, and assess generality across different building types, ventilation regimes, and activities.

• Modeling constraints: The CFD included a reduced mechanism (11 reactions) due to computational expense; comprehensive multiphase and gas-phase chemistry was not fully resolved within CFD and instead provided as constraints from detailed models. - Experimental context: Results are based on a specific test house, an evening bleach application following cooking (layered experiment), with doors open and AHU operation (mixing rate 8 h⁻1; AER during the bleaching period ~0.7 h⁻1), which may limit generalizability to other buildings and conditions. - Assumptions in scale analysis: The spatial/temporal scale framework assumes perturbation-driven steady conditions (e.g., cleaning, cooking); similar gradients may not occur without such perturbations. - Surface processes: Surface interactions and boundary-layer mass transport remain less well characterized and introduce uncertainties in deposition and partitioning predictions. - Measurement spatial coverage: Most species were measured at single locations (e.g., P2) with OH at P7, limiting direct observational mapping of full 3D fields and relying on model-resolved heterogeneity.